Clinical Decision Support

Benefits

A variety of research studies on clinical decision support systems have been conducted and published in the literature. There is a general consensus that clinical decision support technologies have the potential to enhance patient care and at the very least have the potential to modify clinicians' behavior. Clinical reminders and alerts, adherence to treatment plans, and suggested patient education have been reported as effective ways of changing [Page 173]clinician practices. While some may say that these are features that demonstrate the value of clinical decision support, others say that while clinicians' behavior may be shown to be modified, there is little evidence of whether the actual thinking behind the practice modifications is indeed changed. Furthermore, only limited data suggest any improvement in actual patient outcomes. This represents an opportunity for further research and study.

The increasing pressures to monitor and reduce healthcare costs and demonstrate improved outcomes are driving the national trend toward using information as a strategy. Timely data are required to reduce operational inefficiencies and enhance the delivery of patient care. Disparate systems by themselves are inadequate, and data sharing through interfaces presents often inconsistent and conflicting results. Thus, mechanisms are needed to consolidate patient data in a meaningful way to present only the requisite data to make clinical decisions.

Uses

Clinical decision support systems have previously been used for a variety of retrospective analyses. These concepts have expanded into the clinical arena so that data are then presented to clinicians at the point of care and, more important, at the precise time clinical decision making occurs. In its foundation form, the clinical decision support systems include at least one trusted knowledge source (a database of known information about a particular subject, such as drug data) and a set of software programs that establish intelligence (usually referred to as a “rules engine”) to process how the data from the knowledge source may apply to a specific clinical situation. Preestablished rules and guidelines, with corresponding alerts, are developed and edited as necessary by the healthcare organizations. These rules and guidelines typically integrate a variety of clinical data from multiple sources to generate clinician alerts and other treatment suggestions.

Most of these systems have been designed to perform a specific function, such as using data from the knowledge source to validate a medication order for potential drug or therapeutic interactions or against some predetermined range of laboratory result values. Specific rules are established to fit clinical situations, such as if the patient has “X” diagnosis, the “Y and Z” classes of drugs are contraindicated, or if “A” medication is ordered, then the patient must have laboratory values within the range of “B to C.” If the preestablished rule is violated, then an alert is sent back to the prescriber before the order is processed, thus giving the prescriber the opportunity to change the order or asking for an explanation as to why the action is to be taken. Rules and subsequent alerts are usually developed and managed by the healthcare organizations.

...